{"id":"W4210486131","doi":"10.1109/asru51503.2021.9687877","title":"Hybrid Network with Multi-Level Global-Local Statistics Pooling for Robust Text-Independent Speaker Recognition","year":2021,"lang":"en","type":"article","venue":"2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Computer Research Institute of Montréal","funders":"","keywords":"Computer science; Pooling; NIST; Speech recognition; Speaker recognition; Context (archaeology); Artificial intelligence; Set (abstract data type); Speaker diarisation; Hybrid system; Pattern recognition (psychology); Machine learning","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001222225,0.001343361,0.0007988136,0.0006930128,0.0003752137,0.0006018396,0.0013289,0.000814872,0.002708946],"category_scores_gemma":[0.001090215,0.0003893809,0.0007395779,0.0005509001,0.00037618,0.001871255,0.001296087,0.000904327,0.001620814],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004488307,"about_ca_system_score_gemma":0.0004488311,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003486543,"about_ca_topic_score_gemma":0.005258228,"domain_scores_codex":[0.9993829,0.0001133618,0.00003022098,0.0002540043,0.0001338215,0.00008574201],"domain_scores_gemma":[0.9996839,0.00009609335,0.00002949355,0.00005492608,0.0001132421,0.00002252765],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0010448,0.0002673981,0.001399067,0.0002146511,0.0004860168,0.000272733,0.0001842964,0.05351254,0.1757049,0.002512331,0.006729867,0.7576715],"study_design_scores_gemma":[0.00002879122,0.0002721846,0.002135797,0.00001856599,0.0002078889,0.000223693,0.00004092178,0.9297901,0.06101627,0.002269837,0.003931439,0.00006454732],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04235844,0.001784263,0.9480122,0.0002030893,0.0002676388,0.0001008343,0.0003147678,0.004338542,0.002620224],"genre_scores_gemma":[0.654555,0.0007941129,0.3304608,0.0004791669,0.000347242,0.0002470205,0.001443699,0.0003051297,0.01136779],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003486543,"threshold_uncertainty_score":0.00906229,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1497683730595967,"score_gpt":0.283479670430947,"score_spread":0.1337112973713503,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}